Premium
The positive predictive value of “suspicious for high‐grade urothelial carcinoma” in urinary tract cytology specimens: A single‐institution study of 665 cases
Author(s) -
Joudi Anthony M.,
Pambuccian Stefan E.,
Wojcik Eva M.,
Barkan Güliz A.
Publication year - 2016
Publication title -
cancer cytopathology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.29
H-Index - 57
eISSN - 1934-6638
pISSN - 1934-662X
DOI - 10.1002/cncy.21764
Subject(s) - medicine , cytology , urine cytology , bladder cancer , malignancy , biopsy , carcinoma , urothelial carcinoma , urine , carcinoma in situ , urinary system , urology , cancer , pathology
BACKGROUND “The Paris System” proposes a 7‐tier classification system for urine cytology. Establishing the risk of malignancy (ROM) associated with these diagnostic categories is essential to determine the appropriate management of patients. The objective of this study was to determine the ROM associated with the “positive” and “suspicious” categories. METHODS The authors searched their electronic records for urine cytology specimens that had been diagnosed as “positive” or “suspicious” for high‐grade urothelial carcinoma within an 11‐year time frame. Then, the ROM was determined for these specimens within a 6‐month follow‐up interval. The cytologic diagnoses were correlated with surgical biopsy results, follow‐up cytology results, and/or fluorescence in situ hybridization (FISH) results. RESULTS In total, 662 specimens (487 “positive” and 175 “suspicious”), corresponding to 387 patients (295 men and 92 women), were included. The majority of specimens were collected by bladder washing (568 of 662 specimens; 85.4%) and for the indication of surveillance (601 of 662 specimens; 82%). On follow‐up, bladder washing specimens were positive more often positive than voided urine specimens (466 of 570 [81.8%] vs 60 of 92 [65.2%]; P = .0005), and surveillance specimens were more often positive than specimens collected for other indications (82% vs 54.1%). The overall positive predictive value was higher for positive specimens than for suspicious specimens (365 of 461 [79.2%] vs 83 of 150 [55.3%]; P < .0001). CONCLUSIONS Diagnoses of suspicious for high‐grade urothelial carcinoma, as used at the authors' institution, have an ROM that is high but is lower than that for the “positive” category. Therefore, the authors suggest keeping the 2 categories separate, although management should be aggressive in both groups. Cancer Cytopathol 2016;124:811‐9 . © 2016 American Cancer Society .